( PDF ) Rev Osteoporos Metab Miner. 2012; 4 (1) suplemento: 5-16

Gómez de Tejada Romero MJ1, Corral-Gudino L2
1 Departamento de Medicina – Universidad de Sevilla – Sevilla
2 Servicio de Medicina Interna – Hospital el Bierzo – Ponferrada – León

 

Summary
History: Osteoporosis is characterised by a deterioration in bone microarchitecture which puts the bone at risk of suffering fractures. Bazedoxifene is a third generation selective estrogen receptor modulator which has been approved for the treatment of postmenopausal osteoporosis.
Objectives: To evaluate the efficacy of bazedoxifene in the primary and secondary prevention of osteoporotic fractures in postmenopausal women.
Search strategy: Searches were carried out on MEDLINE, Cochrane Central, registers of clinical trials and books of summaries to find random controlled trials published between 2000 and 2011.
Selection criteria: Randomised clinical trials aimed both at the primary and secondary prevention of osteoporosis were selected. Studies were chosen which compared women who received bazedoxifene with those given other drugs for osteoporosis or a placebo.
Compilation and data analysis: The selection of studies and the extraction of data was carried out by two researchers working together. A meta-analysis of the results of fracture and the secondary effects was carried out, establishing the relative risk. The quality of the studies was evaluated on the basis of the criteria proposed by the Cochrane collaboration.
Principal results: Five trials were included in the review (13,543 patients): 3 were concerned with primary prevention (5,622) and two with secondary prevention (7,921). Only those studies of secondary prevention evaluated the fractures as principal objective.
Compared with the placebo bazedoxifene reduced the number of new vertebral fractures detected in the follow up at three years in women with osteoporosis: with a dose of 20 mg the number of patients necessary to treat (NNT) was 56 (CI 95%, 34-146), and at a dose of 40 mg the NNT was 63 (CI 95%, 37-231). In the meta-analysis the relative risk compared with the placebo was 0.59 (CI 95%, 0.44-0.79). There was no difference in the number of symptomatic vertebral fractures or in the number of non-vertebral fractures in the analyses predicted at the start of the study. We found no data on the effect of bazedoxifene on the number of fractures in primary prophylaxis.
For the adverse effects, the meta-analysis did not confirm an increase in the risk of deep vein thrombosis which was seen in the reference study (RR: 8.53). There was an increase in episodes of hot flushes (RR:1.88) or muscular cramps (RR:1.32). No reduction in the incidence of breast cancer was observed, nor in endometrial cancer or endometrial hyperplasia with treatment with bazedoxifene compared with the placebo.
Authors’ conclusions: Bazedoxifene is an efficacious drug in the reduction in the risk of asymptomatic vertebral fractures in primary prophylaxis. In addition, it has been shown to reduce the loss of bone mineral density and to slow bone remodelling in the primary and secondary prevention of osteoporosis. New studies which analyse the risk of non-vertebral fractures and compare the drug with others in the first line of treatment of osteoporosis are necessary in order to understand the true power of their antifractural effect.
Keywords: Osteoporosis. Bone fractures. Selective estrogen receptor modulators. Primary prevention. Secondary prevention.

Introduction

Osteoporosis (OP) is characterised by a deterioration in bone microarchitecture such that it puts the bone at high risk of fractures1. In the absence of curative treatment, up until now therapeutic efforts have been aimed at reducing this risk of fracture by improving the quality and quantity of bone tissue.
Various drugs are being used to this purpose. The objective is to act on bone remodelling with an antiresorptive (such as the biphosphonates, calcitonin) or osteoformative (teriparatide and PTH 1-84) action, or both (strontium ranelate).
The SERMs (selective estrogen receptor modulators) are drugs which have an estrogen agonist/antagonist action depending on the type of hormonal receptors present in each tissue. Their investigation began with the search for a substance with anti-estrogenic action for the prevention and treatment of breast cancer, tamoxifen, which, it was shown, produced an estrogen agonist effect in bone tissue. However, in spite of increasing bone mineral density (BMD)2, it did not reduce the risk of fractures3. In addition, its stimulant action on the endometrium limited its clinical use in OP. Additional advances in molecular biology and pharmacology identified a new compound in this group, raloxifene, considered, therefore, to be a second generation SERM. Its action is estrogen agonist in acting on the receptors of the bone tissue, reducing the resorption favoured by the lack of that hormone (falling within, therefore, the group of drugs with an antiresorptive action), having been demonstrated to increase BMD and reduce the risk of vertebral fractures4, but not with non-vertebral fractures 5. It also exerts this agonist action in the mammary tissue, which is why it is associated with a significant decrease in estrogen-dependent breast cancer6. It improves the lipid profile by reducing blood levels of cholesterol, this being another agonist action7. As an antagonist action, endometrial stimulation does not result7, all of which allows its use in the treatment of postmenopausal OP. However, it has adverse effects, such as an increase in the risk of thromboembolism and leg cramps8, which limit its use. Although it improves the lipid profile, it not been shown to reduce the risk of cardiovascular events9, and is associated with an increase in the vasomotor symptoms of the menopause (breathlessness, flushes, etc)8.
In recent years new compounds have been investigated within the SERM group which may be an improvement on raloxifene: these are what are called the third generation SERMs. Two of these, bazedoxifene10 and lasofoxifene11 have been approved for treatment of postmenopausal OP.
In this work we have carried out a systematic review focusing on those articles published and carried out with bazedoxifene (BZD) for the treatment of postmenopausal OP concerning its tolerance and safety.
The main objective was the evaluation of the efficacy of BZD for the primary and secondary prevention of osteoporotic fractures in postmenopausal women. As secondary objectives its effects on bone mineral density (BMD), the markers for bone remodelling (MBR) and data on the tolerance and safety of the drug were evaluated.

Material and method

The objective of the systematic review was to evaluate the efficacy of BZD in the prevention of new vertebral or non-vertebral fractures in postmenopausal women with or without OP.
The review was carried out in accordance with the PRISMA consensus guidelines for systematic reviews and meta-analyses12.

Selection criteria

Randomised clinical trials were selected which included postmenopausal women. The inclusion was accepted of those trials aimed at both primary and secondary prevention. A comparison was made between groups treated with BZD at any dose and with other drugs for the treatment of OP (biphosphonates, parathyroid hormone, SERM or strontium ranelate) or a placebo. If calcium or vitamin D were used it would have to have been administered to all the treatment groups being compared. As the most appropriate metric for the measurement of the efficacy of the drug, the incidence of vertebral and non-vertebral fractures was evaluated. Changes in levels of BMD and in the MBRs were evaluated as additional results.

Information sources

The studies were identified by means of a search of the PubMed MEDLINE database (January 2006 to June 20011). No language limits were used. The following search strategy was used: ((“osteoporosis, postmenopausal” [MeSH Terms] OR (“osteoporosis” [All Fields] AND “postmenopausal” [All Fields]) OR “postmenopausal osteoporosis” [All Fields] OR “osteoporosis” [All Fields] OR “osteoporosis” [MeSH Terms]) OR (“bone density” [MeSH Terms] OR (“bone” [All Fields] AND “density” [All Fields]) OR “bone density” [All Fields])) AND (“bazedoxifene acetate” [Supplementary Concept] OR “bazedoxifene acetate” [All Fields] OR “bazedoxifene” [All Fields]). Additional studies were found through a search in Cochrane Central and in records of clinical trials (http://www.controlled-trials.com/mrct/) with the term “bazedoxifene”. Lastly, the search was widened to summaries presented at conferences most relevant to the field of osteoporosis: the American Society for Bone and Mineral Research (ASBMR), the European Congress on Osteoporosis and Osteoarthritis (ECCEO), the International Osteoporosis Foundation (IOF) and the American College of Rheumatology (ACR). The period of publication was limited to the period from 2006 to July 2011. The final search of the different sources was carried out on the 21st June 2011.

Selection of studies

Two reviewers (MJ.G.T.R and L.C.G) examined each title and summary generated by the search and identified those articles which were potentially eligible, which were then obtained in full text.
Strategy for quality evaluation
The methodological validity of the articles finally selected was evaluated in accordance with the list of questions proposed in the Cochrane collaboration manual, version 5.1.013. This evaluation consists of the detection of six types of bias: selection (bias due to defects in the generation of the sequence of randomisation and masking of the assignment list), trial performance (bias due to defects in the blinding of the participants in the study and of the health personnel treating them), detection (bias due to defects in the blinding of the evaluation of the results), abandonment or attrition (bias due to the presentation of incomplete results) and information (bias due to selective publication of results). Two reviewers (MJ.G.T.R. and L.C.G) evaluated the quality of each eligible clinical trial, having to indicate whether the risk of bias was high, low, or not possible to evaluate, in accordance with the methodological criteria proposed.
Collection and analysis of data
Two reviewers (MJ.G.T.R. and L.C.G) tabulated the information and data. A summary was created which included information on the methodological aspects of the design of the studies, the characteristics of the participants and the results evaluated (fractures, BMD, MBR, adverse or secondary effects). For the fracture data, all those of which the researchers were informed were considered (whether symptomatic or detected through radiology).
For the results referring to the variation in the incidence of new fractures the necessary number of patients to treat was calculated as well as its confidence interval at 95%. In cases in which when calculating the confidence interval one of the extremes had a negative value, only the lower limit of the confidence interval was established with the positive value calculated 14.
A meta-analysis was carried out of the most significant results using the Review Manager software programme (RevMan) version 5.1, developed by the Nordic Cochrane centre (Copenhagen), of the Cochrane collaboration, 2011. For the results regarding fractures or changes in BMD or in the MBRs, the grouping of the data from postmenopausal patients with OP (secondary prevention) with those of the patients without OP (primary prevention) was not permitted in the meta-analysis. This was due to the fact that it was assumed that the percentages of fractures or the values of bone markers or of mineral density before inclusion in the study would differ between the two groups of women. But the joint meta-analysis of women with or without menopause for drug-related secondary effects was permitted. It was assumed that the presence or absence of OP would not influence those secondary effects derived from the selective blocking of estrogen receptors.

Results

134 articles were found in the systematic literature search. The process of their analysis and selection is described in figure 1. Five articles were selected: one phase 2 clinical trial 15 and four in phase 310,16-18 sponsored by Wyeth (Pfizer). Four of the articles compared BZD10,15-17 and the other a combination of BZD and estrogens combined (EC)18. Three of the articles evaluated the effect in postmenopausal women without OP but with risk factors for its development (primary prevention)16-18. In these three studies, the principal results studied were changes in BMD in the lumbar spine. The other two articles10,15 included women with OP (secondary prevention) and in both, the number of new vertebral fractures was evaluated, in the second as the main result and in the first as a secondary result. The numbers of symptomatic vertebral fractures and of non-vertebral fractures were evaluated in both studies as secondary results. A more detailed study of other characteristics of the design of the studies and of the participants in the clinical trials are given in tables 1 and 2. Apart from the original articles which detail the results in relation to OP, three of the clinical trials had published additional articles with results regarding safety19-23. In two of the clinical trials a sub-analysis was carried out of the results of the study regarding the seriousness24,25 or the characteristics of the patients26 (table 1). The reference study on fractures in secondary prevention was extended by two years27,28. In this prolongation, a complete arm of the study (raloxifene) was suspended.
Figure 2 shows the critical evaluation of the clinical trials according to the methodology proposed in the Cochrane collaboration manual. The item most penalised was that corresponding to the risk of bias due to attrition or loss of patients. This was due to the fact that the abandonment rate was very high (around 30% in the studies at 2 and 3 years), with significant differences in the motives for abandonment between the different groups in one of the studies10,15.

Fractures

Two studies analysed the number of new fractures in patients with OP10,15. In the reference study at 3 years10, the number of new vertebral fractures detected in the radiological follow up at 3 years was lower in those patients who had received 20 mg BZD (NNT, 56; CI 95%, 34-146 patients) or 40 mg BZD (NNT, 63; CI 95%, 37-231 patients) than in those patients who received a placebo. There were no differences with the group which received 60 mg raloxifene (NNT, 56 patients in relation to the placebo; CI 95%, 35-158 patients). There were no differences between the four groups in the number of symptomatic vertebral fractures. The results were similar in the prolongation to 5 years of the original study28. In the study carried out in Japan15 the difference between the rate of new vertebral fractures detected in the radiological follow up at 2 years did not reach statistical significant levels. The meta-analysis of both studies shows a reduction in the risk of vertebral radiological fractures for those groups in treatment with BZD (figure 3).
There were no statistically significant differences in either of the two studies when the numbers of non-vertebral fractures were compared (figure 4).
Post hoc studies were carried out on the reference study at 3 years10 to attempt to define subgroups in which the treatment with BZD was most efficient in the prevention of new fractures. In the original article itself10 there was a sub-study (24% of the patients initially randomised) selecting non-randomly those patients defined as high risk (femoral T-score ≤ -3 or the presence of at least one serious or moderate fracture or minor multiple fractures at the start of the study). This subgroup showed a reduction in non-vertebral fractures in the BZD group compared with raloxifene (NNT, 37; CI 95%, >16 patients) or a placebo (NNT, 29; CI 95%, >15 patients). The same authors, in the prolongation of the original study to 5 years 28, set up a subgroup of 4,216 patients, and, eliminating the branch treated with raloxifene, did not obtain a significant reduction in the appearance of new non-vertebral fractures in high risk patients treated with 20 mg f BZD (37%; p=0.06). In combining the data on both doses, a reduction of 34% (p<0.05) was observed. Kanis et al.24 carried out a post hoc study on the whole sample applying the FRAX tool for the evaluation of the risk of fracture, excluding the branch of patients who took raloxifene. The risk reduction of BZD as opposed to a placebo achieved statistical significance for new vertebral fractures diagnosed in the radiological follow up above the 25th percentile of probability of osteoporotic fracture at 10 years, according to the FRAX scale. For symptomatic vertebral fractures this only reached statistical significance above the 75th percentile.

Bone mineral density

The five studies selected provided data regarding the variation in BMD in different locations (lumbar, hip, femoral neck, trochanter), which was favourable to the treatment groups (BZD, BZD/EC or raloxifene) as against a placebo in all cases (see tables 3 and 4). No differences were found when comparing different doses of BZD. Statistically significant differences were found between BZD and raloxifene in the values of BMD in the hip, in favour of raloxifene in one study10 and, on the contrary, in favour of BZD as against raloxifene in the values of BMD in the lumbar region in some of the treatment groups in the study of the BZD/EC combination (table 4)18. The study by Miller et al.16 did not find differences in the changes in BMD produced by BZD (20 mg and 40 mg) and raloxifene in either of the locations (lumbar and proximal femur).

Makers for bone remodelling

The five studies selected10, 15-18 showed data on the variation in the parameters for bone formation (osteocalcin) and resorption (C-telopeptide). One study15 analysed also the variations on N-telopeptide in blood and in urine. The results were favourable to the treatment groups (BZD, BZD/EC or raloxifene) as against the placebo in all cases. In the comparison of BZD/EC with raloxifene18 the majority of the treatment combinations were superior to raloxifene. There were no differences in the rest of the studies in the comparison of BZD with raloxifene.

Lipid profile

Three studies15-17 analysed the changes in lipid profile. All of these showed a statistically significant reduction in total cholesterol and in LDL cholesterol in the treatment groups (BZD or raloxifene). Some of the studies showed a statistically significant increase in HDL cholesterol in the groups treated with 10 mg and 20 mg of BZD16. In another, the values of lipoprotein (a) was reduced to a statistically significant extent in the groups treated with 20 mg and 40 mg of BZD15.

Adverse effects

Four studies included results regarding the safety of the drug in the original article15-17,19. The analysis of four articles highlighted three most significant and frequent adverse effects in the groups in treatment with BZD: a) episodes of deep vein thrombosis (DVT), with a statistically significant difference in relation to the use of BZD in the study which included the greatest number of patients and the longest follow up period10, but which was not confirmed in the rest of the studies, nor in the meta-analysis (figure 5). The BZD/EC combination did not have a greater frequency of episodes of DVT23; b) the presence of vasodilation and flushing (figure 6); and c) leg cramps (figure 7). In the meta-analysis the last two had a higher risk of occurrence in the group in treatment with BZD as against the placebo. The comparison of the three adverse effects with the raloxifene group had no significant differences10. In the case of vasodilation/flushing the combination of BZD and EC reduced the number of episodes compared with the placebo23.
Other secondary effects were monitored in the cardiovascular area (myocardial infarction, retinal vein thrombosis or cerebral vascular disease), which showed no difference between BZD and placebo.
In terms of data on safety with respect to the female reproductive system, no reduction in the incidence of breast cancer was observed, nor any increase in endometrial cancer or endometrial hyperplasia with the BZD treatment as opposed to the placebo15,17,19,20. In the study of Christiansen et al. the incidence of the diagnosis of fibrocystitis of the breast was significantly higher in the raloxifene group than in the BZD group10,19. The combination of BZD and EC did not produce endometrial hyperplasia. With regard to the reference study at 3 years of secondary prevention of OP, a second clinical trial was registered with a sample of patients on whom had been carried out transvaginal ultrasound. This work showed that treatment with BZD had no effect on changes in endometrial thickness, on the incidence of hyperplasia or endometrial carcinoma, on the presence ovarian cysts or carcinoma, or on the number of episodes of uterine or vaginal haemorrhage.

Discussion

At present, the primordial objective for understanding how efficacious a drug is in the treatment of OP is the reduction in the number of incident fractures. No drug has received this indication without having demonstrated such an efficacy. Of the clinical trials studied, only one had this analysis as its principal objective10. The remaining trials consisted of post hoc studies24,25, or studies in which the incidence of fractures was a secondary objective (the primary objectives being changes in BMD and the MBRs)15-17, or was simply not observed18.
The principal variables of the results evaluated in the remaining clinical trials were changes in BMD and the variation in the MBRs. The results obtained in relation to changes in BMD showed an increase in this measure in all locations with the use of BZD as against the placebo. The comparisons with raloxifene in the different trials showed different results10,16,18. The MBRs behaved in a similar way, with greater reductions in the groups in treatment with BZD. The combination of BZD with EC obtained even better results than raloxifene18. However, when in a clinical trial MBRs or the measurement of BMD are used as its main measure as a substitute for the number of fractures, the results need to be interpreted with care. The relationship between changes in BMD or the MBRs and the reduction in the number of fractures due to antiresorptive treatment is unknown in most cases, and low in those in which it is quantified. The great biological variability which the MBRs possess limit their capacity to individually predict the risk of fracture29. Studies carried out with risedronate have concluded that the changes in the levels of BMD do not predict a reduction in the degree of fractures30,31. The raised incidence of fractures observed in patients with osteopenia corroborates these conclusions32.
In the valuation which the authors themselves make in the study of treatment of OP at three years with BZD using logistic regression indicates that the changes in bone mineral density after 1 and 3 years of treatment with BZD would explain 8% and 29% of hip fractures respectively, or 15% and 43% respectively of fractures in the femoral neck25. The interpretation of changes in the lumbar region does not figure in the study due to methodological problems.
In our review we have not found one on one comparisons of BZD with other drugs recommended as the treatment of choice for OP having demonstrated to reduce the risk of fractures, both vertebral and non-vertebral in randomised clinical trials33: alendronate34, risedronate35, zoledronate36 strontium ranelate37,38 or teriparatide39. Neither at a dose of 20 mg nor at 40 mg did BZD improve the reductive effect on the incidence of vertebral fractures of raloxifene in patients with osteoporosis.
The post hoc analysis of BZD has shown a decrease in non-vertebral fractures in osteoporotic patients at high risk of fracture. However, the interpretation of these results is controversial, since for some authors, the results of post hoc analysis of subgroups should be interpreted with scepticism and should not be used as definitive proof of the effect of a treatment. The recommendation is that all the analyses of subgroups should be planned before carrying out the study to avoid having to look for results which may be statistically significant40.
Similarly to that which happens with raloxifene, its safety level appears to be optimum but special attention should be given to its principal adverse effect, the risk of tromboembolism, whose real incidence is not yet clear. More data are necessary to obtain information on its safety. In addition, the high incidence of cramps and breathlessness may be a significant motive for abandonment. The combination with EC could encourage adhesion to treatment by reducing vasomotor symptoms.
The main limitations for the acceptance of the results of our review in terms of the results obtained are: the finding of only one clinical trial which valued as its main objective that which a priori was the most suitable measure of results; the poor evaluation of the biases in the selected clinical trials, notably the great losses in the follow up and the lack of clarity in the masking of the assignment list. The carrying out of the meta-analysis with the data published and not with the individual patient data is the fundamental limitation in relation to this review. However, our results are similar to those published previously in an independent systematic review of BZD41.
Recently, at the 22nd Congress of the North American Menopause Society, the results obtained from a second extension of two years (total of 7 years of treatment) of the original reference study10 have been presented, in which the raloxifene group has been eliminated and the group on 40 mg of BZD moved to 20 mg, resulting in two branches, 20 mg BZD and placebo. It concluded that in the long term BZD maintained its safety profile and its efficacy in reducing the incidence of vertebral fractures shown in the original study42,43.

Implications for practice

BZD is a drug which is efficacious in the reduction in risk of radiological vertebral fractures in osteoporotic women, but which has not been shown to reduce the number of symptomatic vertebral fractures or non-vertebral fractures. In addition, it is efficacious in reducing the loss of BMD and remodelled bone both in primary and secondary prevention of OP, with a similar action and safety levels as raloxifene. Given that raloxifene is included in the clinical practice guides of the Spanish Society for Bone and Mineral Metabolism Research (SEIOMM), indicated as a first choice drug for menopausal patients under 65 years of age and with a low risk of hip fracture (OP only in the spine, without previous fractures)44, bazedoxifene should be considered in the same position in the therapeutic algorithm. It is possible that it increases the incidence of deep vein thrombosis, but this is not well established.

Implications for research

With respect to the reduction in the incidence of non-vertebral fractures, new clinical trials are necessary whose principal objective is non-vertebral fractures to obtain more conclusive results. In addition, it would be advisable to carry out new studies which compared the use of BZD with a combination of BZD and EC in order to understand which of the two treatments would be most useful in the treatment of OP and which had the better safety profile.

BIBLIOGRAPHY
1. Osteoporosis prevention, diagnosis, and therapy. JAMA 2001;285:785-95.
2. Love RR, Mazess RB, Barden HS, Epstein S, Newcomb PA, Jordan VC, et al. Effects of tamoxifen on bone mineral density in postmenopausal women with breast cancer. N Engl J Med 1992;326:852-6.
3. Vestergaard P, Rejnmark L, Mosekilde L. Effect of tamoxifen and aromatase inhibitors on the risk of fractures in women with breast cancer. Calcif Tissue Int 2002;82:334-40.
4. Ettinger B, Black DM, Mitlak BH, Knickerbocker RK, Nickelsen T, Genant HK, et al. Reduction of vertebral fracture risk in postmenopausal women with osteoporosis treated with raloxifene: results from a 3-year randomized clinical trial. Multiple Outcomes of Raloxifene Evaluation (MORE) Investigators. JAMA 1999;282:637-45.
5. Cranney A, Tugwell P, Zytaruk N, Robinson V, Weaver B, Adachi J, et al. Meta-analyses of therapies for postmenopausal osteoporosis. IV. Meta-analysis of raloxifene for the prevention and treatment of postmenopausal osteoporosis. Endocr Rev 2002;23:524-8.
6. Nelson HD, Fu R, Griffin JC, Nygren P, Smith ME, Humphrey L. Systematic review: comparative effectiveness of medications to reduce risk for primary breast cancer. Ann Intern Med 2009;151:703-15.
7. Delmas PD, Bjarnason NH, Mitlak BH, Ravoux AC, Shah AS, Huster WJ, et al. Effects of raloxifene on bone mineral density, serum cholesterol concentrations, and uterine endometrium in postmenopausal women. N Engl J Med 1997;337:1641-7.
8. Delmas PD, Ensrud KE, Adachi JD, Harper KD, Sarkar S, Gennari C, et al. Efficacy of raloxifene on vertebral fracture risk reduction in postmenopausal women with osteoporosis: four-year results from a randomized clinical trial. J Clin Endocrinol Metab 2002;87:3609-17.
9. Barrett-Connor E, Mosca L, Collins P, Geiger MJ, Grady D, Kornitzer M, et al. Effects of raloxifene on cardiovascular events and breast cancer in postmenopausal women. N Engl J Med 2006;355:125-37.
10. Silverman SL, Christiansen S, Genant HK, Vukicevic S, Zanchetta JR, de-Villiers TJ, et al. Efficacy of bazedoxifene in reducing new vertebral fracture risk in postmenopausal women with osteoporosis: results from a 3-year, randomized, placebo-, and active-controlled clinical trial. J Bone Miner Res 2008;23:1923-34.
11. Cummings SR, Ensrud K, Delmas PD, LaCroix AZ, Vukicevic S, Reid DM, et al. Lasofoxifene in postmenopausal women with osteoporosis. N Engl J Med 2010;362:686-96.
12. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ 2009;339:b2535.
13. The Cochrane Collaboration. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011). Higgins JPT and Green S. www.cochrane-handbook.org. 2011. Ref Type: Electronic Citation.
14. Altman DG. Confidence intervals for the number needed to treat. BMJ 1998;317:1309-12.
15. Itabashi A, Yoh K, Chines AA, Miki T, Takada M, Sato H, et al. Effects of bazedoxifene on bone mineral density, bone turnover, and safety in postmenopausal Japanese women with osteoporosis. J Bone Miner Res 2011;26:519-29.
16. Miller PD, Chines AA, Christiansen C, Hoeck HC, Kendler DL, Lewiecki EM, et al. Effects of bazedoxifene on BMD and bone turnover in postmenopausal women: 2-yr results of a randomized, double-blind, placebo-, and active-controlled study. J Bone Miner Res 2008;23:525-35.
17. Xu L, Tsai KS, Kim GS, Wu Y, Vincendon P, Chines AA, et al. Efficacy and safety of bazedoxifene in postmenopausal Asian women. Osteoporos Int 2011;22:559-65.
18. Lindsay R, Gallagher JC, Kagan R, Pickar JH, Constantine G. Efficacy of tissue-selective estrogen complex of bazedoxifene/conjugated estrogens for osteoporosis prevention in at-risk postmenopausal women. Fertil Steril 2009;92:1045-52.
19. Christiansen C, Chesnut CH, Adachi JD, Brown JP, Fernandes CE, Kung AW, et al. Safety of bazedoxifene in a randomized, double-blind, placebo- and active-controlled Phase 3 study of postmenopausal women with osteoporosis. BMC. Musculoskelet Disord 2010;11:130.
20. Pinkerton JV, Archer DF, Utian WH, Menegoci JC, Levine AB, Chines AA, et al. Bazedoxifene effects on the reproductive tract in postmenopausal women at risk for osteoporosis. Menopause 2009;16:1102-8.
21. Archer DF, Pinkerton JV, Utian WH, Menegoci JC, de-Villiers TJ, Yuen CK, et al. Bazedoxifene, a selective estrogen receptor modulator: effects on the endometrium, ovaries, and breast from a randomized controlled trial in osteoporotic postmenopausal women. Menopause 2009;16:1109-15.
22. Pickar JH, Yeh IT, Bachmann G, Speroff L. Endometrial effects of a tissue selective estrogen complex containing bazedoxifene/conjugated estrogens as a menopausal therapy. Fertil Steril 2009;92:1018-24.
23. Lobo RA, Pinkerton JV, Gass ML, Dorin MH, Ronkin, Pickar JH, et al. Evaluation of bazedoxifene/conjugated estrogens for the treatment of menopausal symptoms and effects on metabolic parameters and overall safety profile. Fertil Steril 2009;92:1025-38.
24. Kanis JA, Johansson H, Oden A, McCloskey EV. Bazedoxifene reduces vertebral and clinical fractures in postmenopausal women at high risk assessed with FRAX. Bone 2009;44:1049-54.
25. Bruyere O, Detilleux J, Chines A, Reginster JY. Relationships between changes in bone mineral density or bone turnover markers and vertebral fractures incidence in patients treated with bazedoxifene. Osteoporos Int 2011;23 Suppl 1. Ref Type: Abstract.
26. Levine AB, Ciesielska M, Chines A. Efficacy and Safety of Bazedoxifene in Postmenopausal African American Women. J Bone Miner Res 2010;25 Suppl 1. Ref Type: Abstract.
27. De Villiers TJ, Chines AA, Palacios S, Lips P, Sawicki AZ, Levine AB, et al. Safety and tolerability of bazedoxifene in postmenopausal women with osteoporosis: results of a 5-year, randomized, placebo-controlled phase 3 trial. Osteoporos Int 2011;22:567-76.
28. Silverman SL, Chines AA, Kendler DL, Kung AW, Teglbjaerg CS, Felsenberg D, et al. Sustained efficacy and safety of bazedoxifene in preventing fractures in postmenopausal women with osteoporosis: results of a 5-year, randomized, placebo-controlled study. Osteoporos Int 2011; Jul 21 [Epub ahead of print].
29. Civitelli R, Armamento-Villareal R, Napoli N. Bone turnover markers: understanding their value in clinical trials and clinical practice. Osteoporos Int 2009;20:843-51.
30. Watts NB, Geusens P, Barton IP, Felsenberg D. Relationship between changes in BMD and nonvertebral fracture incidence associated with risedronate: reduction in risk of nonvertebral fracture is not related to change in BMD. J Bone Miner Res 2005;20:2097-104.
31. Watts NB, Cooper C, Lindsay R, Eastell R, Manhart MD, Barton IP, et al. Relationship between changes in bone mineral density and vertebral fracture risk associated with risedronate: greater increases in bone mineral density do not relate to greater decreases in fracture risk. J Clin Densitom 2004;7:255-61.
32. Arboleya L, Díaz-Curiel M, Del Río L, Blanch J, Díez-Pérez A, Guañabens N, et al. Prevalence of vertebral fracture in postmenopausal women with lumbar osteopenia using MorphoXpress® (OSTEOXPRESS Study). Aging Clin Exp Res 2010;22:419-26.
33. National Institute for Health and Clinical Excellence (NICE). Alendronate, etidronate, risedronate, raloxifene, strontium ranelate and teriparatide for the secondary prevention of osteoporotic fragility fractures in postmenopausal women. NHS evidence. http://guidance.nice.org.uk/TA161. 2011. 8-9-2011. Ref Type: Electronic Citation.
34. Cranney A, Wells G, Willan A, Griffith L, Zytaruk N, Robinson V, et al. Meta-analyses of therapies for postmenopausal osteoporosis. II. Meta-analysis of alendronate for the treatment of postmenopausal women. Endocr Rev 2002;23:508-16.
35. Cranney A, Tugwell P, Adachi J, Weaver B, Zytaruk N, Papaioannou A, et al. Meta-analyses of therapies for postmenopausal osteoporosis. III. Meta-analysis of risedronate for the treatment of postmenopausal osteoporosis. Endocr Rev 2002;23:517-23.
36. Black DM, Delmas PD, Eastell R, Reid R, Boonen S, Cauley JA, et al. Once-yearly zoledronic acid for treatment of postmenopausal osteoporosis. N Engl J Med 2007;356:1809-22.
37. Reginster JY, Seeman E, De Vernejoul MC, Adami S, Compston J, Phenekos C, et al. Strontium ranelate reduces the risk of nonvertebral fractures in postmenopausal women with os